Neurosymbolic Reasoning Shortcuts under the Independence Assumption

Emile van Krieken, Pasquale Minervini, Edoardo Ponti, Antonio Vergari
Proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning, PMLR 284:285-302, 2025.

Abstract

The ubiquitous independence assumption among symbolic concepts in neurosymbolic (NeSy) predictors is a convenient simplification: NeSy predictors use it to speed up probabilistic reasoning. Recent works like van Krieken et al. (2024) and Marconato et al. (2024) argued that the independence assumption can hinder learning of NeSy predictors and, more crucially, prevent them from correctly modelling uncertainty. There is, however, scepticism in the NeSy community around the scenarios in which the independence assumption actually limits NeSy systems (Faronius and Dos Martires, 2025). In this work, we settle this question by formally showing that assuming independence among symbolic concepts entails that a model can never represent uncertainty over certain concept combinations. Thus, the model fails to be aware of _reasoning shortcuts_, i.e., the pathological behaviour of NeSy predictors that predict correct downstream tasks but for the wrong reasons.

Cite this Paper

BibTeX
@InProceedings{pmlr-v284-krieken25a, title = {Neurosymbolic Reasoning Shortcuts under the Independence Assumption}, author = {Krieken, Emile van and Minervini, Pasquale and Ponti, Edoardo and Vergari, Antonio}, booktitle = {Proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning}, pages = {285--302}, year = {2025}, editor = {H. Gilpin, Leilani and Giunchiglia, Eleonora and Hitzler, Pascal and van Krieken, Emile}, volume = {284}, series = {Proceedings of Machine Learning Research}, month = {08--10 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v284/main/assets/krieken25a/krieken25a.pdf}, url = {https://proceedings.mlr.press/v284/krieken25a.html}, abstract = {The ubiquitous independence assumption among symbolic concepts in neurosymbolic (NeSy) predictors is a convenient simplification: NeSy predictors use it to speed up probabilistic reasoning. Recent works like van Krieken et al. (2024) and Marconato et al. (2024) argued that the independence assumption can hinder learning of NeSy predictors and, more crucially, prevent them from correctly modelling uncertainty. There is, however, scepticism in the NeSy community around the scenarios in which the independence assumption actually limits NeSy systems (Faronius and Dos Martires, 2025). In this work, we settle this question by formally showing that assuming independence among symbolic concepts entails that a model can never represent uncertainty over certain concept combinations. Thus, the model fails to be aware of _reasoning shortcuts_, i.e., the pathological behaviour of NeSy predictors that predict correct downstream tasks but for the wrong reasons.} }
Endnote
%0 Conference Paper %T Neurosymbolic Reasoning Shortcuts under the Independence Assumption %A Emile van Krieken %A Pasquale Minervini %A Edoardo Ponti %A Antonio Vergari %B Proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning %C Proceedings of Machine Learning Research %D 2025 %E Leilani H. Gilpin %E Eleonora Giunchiglia %E Pascal Hitzler %E Emile van Krieken %F pmlr-v284-krieken25a %I PMLR %P 285--302 %U https://proceedings.mlr.press/v284/krieken25a.html %V 284 %X The ubiquitous independence assumption among symbolic concepts in neurosymbolic (NeSy) predictors is a convenient simplification: NeSy predictors use it to speed up probabilistic reasoning. Recent works like van Krieken et al. (2024) and Marconato et al. (2024) argued that the independence assumption can hinder learning of NeSy predictors and, more crucially, prevent them from correctly modelling uncertainty. There is, however, scepticism in the NeSy community around the scenarios in which the independence assumption actually limits NeSy systems (Faronius and Dos Martires, 2025). In this work, we settle this question by formally showing that assuming independence among symbolic concepts entails that a model can never represent uncertainty over certain concept combinations. Thus, the model fails to be aware of _reasoning shortcuts_, i.e., the pathological behaviour of NeSy predictors that predict correct downstream tasks but for the wrong reasons.
APA
Krieken, E.v., Minervini, P., Ponti, E. & Vergari, A.. (2025). Neurosymbolic Reasoning Shortcuts under the Independence Assumption. Proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning, in Proceedings of Machine Learning Research 284:285-302 Available from https://proceedings.mlr.press/v284/krieken25a.html.

Related Material